Currently the processing of biochemistry samples has a number of key drawbacks. These include the volume size—resulting in high reagent costs; high consumable costs; and labour-intensive protocols and processes which are highly susceptible to cross-contamination. For these reasons complete control and isolation of each individual sample within the biochemistry process cannot currently be ensured.
For a number of biochemistry process applications—sequence bead preparation, pyrosequencing, nucleic acid ligation, and polymerase chain reaction—and not limited to these, the limitations of volume size, chemistry cost, labour cost, and the reaction efficiency are evident.
Sequence bead preparation is a process by which small beads are coated in an application-specific chemistry. For example in DNA replication, the beads are coated initially with DNA primers in advance of the amplification process. Even for today's state-of-the-art sequencers a relatively high local concentration of the target molecule is required to sequence accurately. Current estimates for a typical protocol estimate that only 80% of the beads processed are sufficiently coated to ensure accurate sequencing. Therefore to ensure a relatively high concentration of the target sample a large number of beads must be used for statistical accuracy. Furthermore, the transferral of even coated beads from one well to another inevitably leads to losses of both the beads and the suspended fluid. This is a result of dead volumes and inefficiencies inherent in today's pipetting and liquid handling systems. This biochemistry process is generally performed in 96 or 384 static well plates with typical volumes ranging from 10 microliters to 200 microliters.
Another biochemistry process, pyrosequencing, mixes a relatively high concentration of nucleic acid with primer-coated beads. The nucleic acids attach and form a clonal colony on the beads. This is then amplified using emulsion-based PCR. The sequencing machine contains a large number of picoliter-volume wells that are large enough for a single bead along with the relevant enzymes required for sequencing. Pyrosequencing uses the luciferase enzyme to generate light as read-out, and the sequencing machine takes a picture of the wells for every added nucleotide. One of the key difficulties in this process is the efficient coating of the beads with primers. A percentage of beads using current technologies are not properly coated with primer chemistry, resulting in poorer reaction efficiencies. Using today's technologies to improve the coating efficiencies of the beads would require an unsustainable increase in reagent cost.
Within nucleic acid ligation similar biochemistry processing issues arise. Nucleic acid ligation has become an important tool in modem molecular biology research for generating recombinant nucleic acid sequences. For example, nucleic acid ligases are used with restriction enzymes to insert nucleic acid fragments, often genes, into plasmids for use in genetic engineering. Nucleic acid ligation is a relatively common technique in molecular biology wherein short strands of DNA may be joined together by the action of an enzyme called ligase at a specific temperature, commonly 16-25° C. depending on the protocol used. To join more than two sequences of short DNA strands together, for example in the construction of a synthetic genetic sequence, it is impossible to combine all the DNA strands and then perform the ligation. This would result in random sequences in which the end of one strand would be joined to the start of an incorrect strand. This incorrect sequence, or orientation, would not be desirable in a synthetically-constructed gene where the order of the genetic code is crucial. To perform the technique correctly pairwise combinations of neighbouring sequences must first be ligatated to yield the correct orientation. These paired synthetic constructs may then be ligated in the correct orientation to yield even longer synthetic constructs. The process involves a large and intricate amount of chemistry processing and manipulation. This can be quite a labour intensive process or if performed using today's liquid handling and results in large consumable costs and suffers from the known dead volume losses of the static well plates and pipette aspirations. Also using today's liquid handling technologies the mixing and control of small volumes is limited by the ability to aspirate and manipulate relatively small volumes. Typical volumes used in nucleic acid ligation are 10-200 microliters with nucleic acid strand lengths between 50-200 base pairs.
Polymerase Chain Reaction (PCR) has been used extensively to amplify targeted DNA and cDNA for many applications in molecular biology. The PCR technique amplifies a single or a few copies of a piece of DNA, generating thousands to billions of copies of a particular DNA sequence. Modem PCR instruments carry out the PCR process in reaction volumes ranging from 10-200 micro-liters. One of the largest obstacles to carrying out PCR in small volumes is the difficulty in manipulating small volumes of the constituent reagents with manual pipettes. The large volume size is a direct result of the poor capability of existing technologies to dispense and mix sub-nanoliter volumes. Furthermore, for the next generation microfluidic technologies based on flowing systems, these are still limited by the starting volume dispensed versus the actual amount of sample required for the biochemistry process. These microfluidic systems are also limited during the biochemistry process to a defined protocol control of the samples. These systems typically rely on micro-scale fluid channel networks to transport and mix sub-microliter volumes. Some of the major drawbacks of these technologies are: the single use of the microfluidic cards—to prevent contamination—the lack of dynamic control of the each individual sample—transporting and mixing any individual sample at any point in the biochemistry process—and the closed architecture of the system.
In particular, current methods of Digital Polymerase Chain Reaction (dPCR) are performed through the division of an initial sample into multiple smaller volumes samples until one DNA template remains in each subvolume. Counting the number of positive subvolumes which contain DNA, the starting copy number in the original volume can be calculated. Typically, this involves multiple serial dilution steps to generate a sample volume with statistically one DNA target per reaction volume. Statistically a subset of the total volume may be tested to determine the initial copy number, allowing for a reduction in the total number of PCR reactions. However for rare target detection a larger subset of volumes need to be tested to improve the statistical accuracy. This results in a larger number of blank volumes and a larger test volume—resulting in the use of more chemistry, time, instrumentation, sample handling, and processing steps.
Another method of dPCR is whereby an emulsion of the test volume is generated in an oil-based carrier. This method is an effort to reduce the number of instruments required and time required for a result. First, the target sample is diluted and emulsified into small enough volumes with a statistical distribution of less than one copy per droplet, within the carrier oil. This larger volume can then be treated as a single sample volume and processed using PCR protocols. However this method is generally limited to end point detection. Further instrumentation is required in the form of a flow cytometer, thereby being able to detect the target presence per droplet flowing past a sensor. Flow cytometers are low speed; expensive; can require specific fluid mediums and only allow for endpoint detection. The limitations of endpoint detection include the requirement of a post processing step; lower sensitivity;
longer time to result; specificity and more instrumentation. A further challenge for emulsion based PCR methods is the stability required and control of each droplet. Droplet merging or splitting introduces further statistical errors into the processing.
Today's pipetting and liquid handling systems are unable to process 100% of the given starting volume. For pipettes both the liquid storage system—static well plates—and the mechanical actuation within the system prevent complete aspiration of the sample. This loss or dead volume in static plates can be accounted for by the surface wetting characteristics and the geometry, neither of which current technologies can account for.
In flowing systems the collection of individual biological samples during or at the end of the biochemistry process is proving to be very challenging for existing technologies. The typical continuous flowing systems comprise of pumps and reservoirs which generally make the easy retrieval of critical fluids, particularly at the microscale, technically difficult. Also, within flowing systems initial priming of the system is time consuming, costly and if done incorrectly leads to a catastrophic failure of the test requiring a retest of the biological sample.
Another drawback to existing biochemistry processing is the inability to automate the biochemistry process for nano-liter and sub-nano-liter volumes. The transport, mixing or retrieval of each individual sample cannot be performed by existing automated technologies.
In more general chemistry processing, such as generic microchemistry, where the manipulation of small amounts of fluid is necessary, one can clearly see the limitations of current technology in the volume waste fluid remaining in the static well plates or within the system. This is a result of current technology's lack of capability to dispense and control smaller volumes demanded by evermore sophisticated molecular biology techniques, and the call for improved efficiencies.
The invention is therefore directed towards providing improved sample handling to overcome at least some of the above problems.